Data Science Center »

Refining data search strategy

At a Glance

Quick Links

Archive Tutorials

Have questions about your research?
We can help!

Refining Research

You probably have in mind a general area or research topic you want to pursue. When you talk with one of the Archive staff we will help you further refine you ideas by narrowing in on four areas: Geography, Time, Population and Context.

Geography

Let's say your topic is about Earthquakes.Think about what geographic areas you'd like to cover with this topic. You should consider this in terms of the following:

 

  • Physical Areas

Fault zones, mountain regions, water areas, etc. 

  • Governmental Units:

Cities, counties, states, countries 

  • Political Boundaries:

Congressional districts 

  • Other Units:

School districts, hospital or fire zones, telephone service areas, water districtsNote: The Archive has reference sources to use in understanding geographic units and boundaries.

 

Time

Consider whether your research area covers recent events, or is historical or look at changes over a specified range of time. 

 

  • Current: Many data files are about a year old before they are released to the public. If you need very recent information, you will need to use other sources, such as newspapers and journal articles with tables or other quantitative information rather than raw data.
  • Historical: Computers have been popularly used in survey research since the 1960's. If you need data for earlier time periods it may be more difficult to locate in an electronic format.




Population

What is the population or case group you wish to study? This can be a group or groups of people, particular events, official records, etc. In addition you should consider whether you will look at a specific sample or subset of people, events, records, etc.



Examples :

  • You may be able to describe a sample of people buy specifying gender, race, age, ethnicity, etc.
  • An example of events might be occurrences of earthquakes. You can describe a sample of these occurrences by looking at only those registering magnitude 5 and above on the Richter scale.
  • Within the universe of all official records, you might want to look oat only birth certificates, perhaps for only those issued for children born to mothers under 15 years of age.




Context

Raw data can be used in a variety of ways. Continuing with the earthquake example, even after you have specified the geographic region, the time period, the population and sample you desire, you will want to look at the data from a certain perspective, or within a context relevant to your area of study. This context will determine the type of data most usefulfor analysis.

For example, if you want to study people who have experienced high magnitude earthquakes in Los Angeles in the last 20 years, do you want to look at how quality of life is affected? Do you want to examine scales to determine emotional response to earthquakes? Do you want to study the relationship of quality of life and social or financial resources?